It is proposed herein to prepare the first family of functional, selective, cation-conducting channel molecules that can insert into a lipid bilayer and mediate cation flux. Individual attempts have been made to prepare such compounds in the past, but our proposed systematic and "modular" approach will permit us to examine the requirements of lipophilicity, polarity, distance, channel diameter, flexibility, the role of water, etc. The basic design of the channel proposed here is a two-compartment structure that mimics the dimer structure of gramicidin, a natural channel- forming peptide. The polar head groups and central anchor will be macrocyclic polyethers and the "sides" of the molecular compartments will be shaped and rigidified by steroids. Steroidal crown ethers have been shown in previous studies to form vesicles and steroids are known to be the natural rigidifying element of many cellular and subcellular bilayer membranes. Indeed, two macrocyclic rings separated by only 12 carbon atoms have now been shown to form vesicles as well. Flexible sidearms opposite the steroids will provide the other wall of the channel and may serve the dual purpose of intra-channel cation relay. Whether such relays, in addition to the central macroring, will be required for effective cation transport is currently unknown and must be determined by experiment. The flexible sidearms, steroids, or crowns may associate with that could perform the relay function. Such questions as the requirements for rigidity, polar groups, cation jumping distances, etc., will be addressed by the preparation of several discrete molecular compartments or boxes. These systems are of interest in their own right as possible receptors for cations and small molecules. Study of these smaller systems will permit a more rational design of the molecular channel.